Top Banner
This journal is c the Owner Societies 2012 Phys. Chem. Chem. Phys., 2012, 14, 675–680 675 Cite this: Phys. Chem. Chem. Phys., 2012, 14, 675–680 Photodissociation of isobutene at 193 nm Gabriel M. P. Just, ab Bogdan Negru, ab Dayoung Park ab and Daniel M. Neumark* ab Received 19th August 2011, Accepted 28th October 2011 DOI: 10.1039/c1cp22651g The collisionless photodissociation dynamics of isobutene (i-C 4 H 8 ) at 193 nm via photofragment translational spectroscopy are reported. Two major photodissociation channels were identified: H+C 4 H 7 and CH 3 + CH 3 CCH 2 . Translational energy distributions indicate that both channels result from statistical decay on the ground state surface. Although the CH 3 loss channel lies 13 kcal mol 1 higher in energy, the CH 3 : H branching ratio was found to be 1.7 (5), in reasonable agreement with RRKM calculations. I. Introduction Isobutene, i-C 4 H 8 , (2-methylpropene) is the smallest branched alkene. It plays a key role in combustion chemistry as an intermediate in the pyrolysis of iso-octane and in the oxidation of fuel additives such as MTBE and ETBE (methyl and ethyl t-butyl ether). 1 The chemistry of isobutene in the Earth’s troposphere, notably its reactions with NO 2 and NO 3 , 2,3 is of interest, as are its reactions with free radicals in the atmosphere of Titan in order to form larger hydrocarbons. 4 Isobutene has been implicated as a product in the O( 3 P)+t-C 4 H 9 (t-butyl) radical–radical reaction 5 and from the photodissociation of tert-C 4 H 9 . 6,7 However, the photodissociation of isobutene itself has not been reported previously. In this paper, we investigate the collisionless photodissociation of isobutene at 193 nm in order to gain new insights into its unimolecular photochemistry and dissociation dynamics. The work presented here is motivated by numerous studies of the bimolecular and unimolecular kinetics of isobutene in shock tubes and flames. 8–13 These studies have focused on elucidating the mechanisms for the oxidation and pyrolysis of isobutene. An issue arising from this body of work is the identity of the products arising from the unimolecular decay of isobutene. In some kinetics studies, reaction mechanisms are proposed in which only H atom loss is included, 10,12 whereas others also include the somewhat higher energy CH 3 loss channel. 8,13,14 Photodissocia- tion measurements provide unambiguous identification of the primary products from photoexcitation to an excited electronic state. In cases where dissociation proceeds via internal conversion to the ground state followed by statistical decay, the results of photodissociation experiments can have direct bearing on the interpretation of kinetics experiments in which it is often very difficult to identify product channels for specific reactions. Photodissociation of isobutene is also of interest in light of previous work by Zierhut et al. 6 and our group 7 on the photodissociation of the t-C 4 H 9 radical near 248 nm. One concern in those experiments was that some observed channels were from the photodissociation of vibrationally hot isobutene produced in the pyrolysis source used to generate t-butyl radical rather than from t-butyl itself. An independent study of isobutene photodissociation could thus corroborate the interpretation of the previous experiments on t-butyl. The UV absorption spectrum of the isobutene molecule begins around 205 nm and comprises numerous closely spaced bands; 15,16 the band around 193 nm has been assigned to the lowest-lying pp* transition. 17,18 Excitation at 193 nm can lead to photodissociation by two bond cleavage channels involving loss of either an H atom or a CH 3 group: (CH 3 ) 2 CCH 2 + hn - CH 2 CH 3 CCH 2 +H DH 0 = 88.3 kcal mol 1 , 19 (1) (CH 3 ) 2 CCH 2 + hn - CH 3 CCH 2 + CH 3 DH 0 = 100.9 kcal mol 1 , (2) The 2-methylallyl radical from channel 1, can further dissociate to form allene (C 3 H 4 ) via the loss of a methyl group. 20 The barrier to this dissociation process has been calculated by Li et al. 21 to be 55.5 kcal mol 1 . Furthermore, the 2-propenyl radical from channel 2 can undergo a 1,2-hydrogen shift over an isomerization barrier of 45.4 kcal mol 1 to form the allyl radical. 22 Fig. 1 shows the primary energetics for channels 1 and 2 as well as the barrier heights and energies for subsequent dissocia- tion and isomerization. In this work, we investigate the collisionless photodissociation of isobutene at 193 nm using molecular beam photofragment translational spectroscopy. This experiment yields the kinetic energy and angular distribution for each photofragmentation channel, enabling the direct identification of the primary photo- fragments and providing insight into the dissociation mechanism. a Department of Chemistry, University of California, Berkeley, CA 94720, USA. E-mail: [email protected] b Chemical Sciences Division, Lawrence Berkeley National Laboratory, Berkeley, CA 94720, USA PCCP Dynamic Article Links www.rsc.org/pccp PAPER Downloaded by University of California - Berkeley on 17 January 2012 Published on 28 November 2011 on http://pubs.rsc.org | doi:10.1039/C1CP22651G View Online / Journal Homepage / Table of Contents for this issue
6

Citethis:Phys. Chem. Chem. Phys.,2012,14 ,675680 PAPER · his ournal is c the Owner Societies 2012 Phys. Chem. Chem. Phys.,2012,14,675680 677 Many more TOF spectra were taken at various

Mar 22, 2020

Download

Documents

dariahiddleston
Welcome message from author
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Page 1: Citethis:Phys. Chem. Chem. Phys.,2012,14 ,675680 PAPER · his ournal is c the Owner Societies 2012 Phys. Chem. Chem. Phys.,2012,14,675680 677 Many more TOF spectra were taken at various

This journal is c the Owner Societies 2012 Phys. Chem. Chem. Phys., 2012, 14, 675–680 675

Cite this: Phys. Chem. Chem. Phys., 2012, 14, 675–680

Photodissociation of isobutene at 193 nm

Gabriel M. P. Just,ab

Bogdan Negru,ab

Dayoung Parkab

and

Daniel M. Neumark*ab

Received 19th August 2011, Accepted 28th October 2011

DOI: 10.1039/c1cp22651g

The collisionless photodissociation dynamics of isobutene (i-C4H8) at 193 nm via photofragment

translational spectroscopy are reported. Two major photodissociation channels were identified:

H + C4H7 and CH3 + CH3CCH2. Translational energy distributions indicate that both channels

result from statistical decay on the ground state surface. Although the CH3 loss channel lies

13 kcal mol�1 higher in energy, the CH3 :H branching ratio was found to be 1.7 (5), in

reasonable agreement with RRKM calculations.

I. Introduction

Isobutene, i-C4H8, (2-methylpropene) is the smallest branched

alkene. It plays a key role in combustion chemistry as an

intermediate in the pyrolysis of iso-octane and in the oxidation

of fuel additives such as MTBE and ETBE (methyl and ethyl

t-butyl ether).1 The chemistry of isobutene in the Earth’s

troposphere, notably its reactions with NO2 and NO3,2,3 is of

interest, as are its reactions with free radicals in the atmosphere

of Titan in order to form larger hydrocarbons.4 Isobutene has

been implicated as a product in the O(3P)+t-C4H9(t-butyl)

radical–radical reaction5 and from the photodissociation of

tert-C4H9.6,7 However, the photodissociation of isobutene itself

has not been reported previously. In this paper, we investigate

the collisionless photodissociation of isobutene at 193 nm in

order to gain new insights into its unimolecular photochemistry

and dissociation dynamics.

The work presented here is motivated by numerous studies of

the bimolecular and unimolecular kinetics of isobutene in shock

tubes and flames.8–13 These studies have focused on elucidating

the mechanisms for the oxidation and pyrolysis of isobutene. An

issue arising from this body of work is the identity of the products

arising from the unimolecular decay of isobutene. In some

kinetics studies, reaction mechanisms are proposed in which only

H atom loss is included,10,12 whereas others also include the

somewhat higher energy CH3 loss channel.8,13,14 Photodissocia-

tion measurements provide unambiguous identification of the

primary products from photoexcitation to an excited electronic

state. In cases where dissociation proceeds via internal conversion

to the ground state followed by statistical decay, the results of

photodissociation experiments can have direct bearing on the

interpretation of kinetics experiments in which it is often very

difficult to identify product channels for specific reactions.

Photodissociation of isobutene is also of interest in light

of previous work by Zierhut et al.6 and our group7 on the

photodissociation of the t-C4H9 radical near 248 nm. One

concern in those experiments was that some observed channels

were from the photodissociation of vibrationally hot isobutene

produced in the pyrolysis source used to generate t-butyl

radical rather than from t-butyl itself. An independent study

of isobutene photodissociation could thus corroborate the

interpretation of the previous experiments on t-butyl.

The UV absorption spectrum of the isobutene molecule

begins around 205 nm and comprises numerous closely spaced

bands;15,16 the band around 193 nm has been assigned to the

lowest-lying pp* transition.17,18 Excitation at 193 nm can lead

to photodissociation by two bond cleavage channels involving

loss of either an H atom or a CH3 group:

(CH3)2CCH2 + hn - �CH2CH3CCH2 + H

DH0 = 88.3 kcal mol�1,19 (1)

(CH3)2CCH2 + hn - CH3C�CH2 + CH3

DH0 = 100.9 kcal mol�1, (2)

The 2-methylallyl radical from channel 1, can further dissociate

to form allene (C3H4) via the loss of a methyl group.20 The

barrier to this dissociation process has been calculated by

Li et al.21 to be 55.5 kcal mol�1. Furthermore, the 2-propenyl

radical from channel 2 can undergo a 1,2-hydrogen shift over

an isomerization barrier of 45.4 kcal mol�1 to form the allyl

radical.22 Fig. 1 shows the primary energetics for channels 1 and 2

as well as the barrier heights and energies for subsequent dissocia-

tion and isomerization.

In this work, we investigate the collisionless photodissociation

of isobutene at 193 nm using molecular beam photofragment

translational spectroscopy. This experiment yields the kinetic

energy and angular distribution for each photofragmentation

channel, enabling the direct identification of the primary photo-

fragments and providing insight into the dissociation mechanism.

aDepartment of Chemistry, University of California, Berkeley,CA 94720, USA. E-mail: [email protected]

bChemical Sciences Division, Lawrence Berkeley National Laboratory,Berkeley, CA 94720, USA

PCCP Dynamic Article Links

www.rsc.org/pccp PAPER

Dow

nloa

ded

by U

nive

rsity

of

Cal

ifor

nia

- B

erke

ley

on 1

7 Ja

nuar

y 20

12Pu

blis

hed

on 2

8 N

ovem

ber

2011

on

http

://pu

bs.r

sc.o

rg |

doi:1

0.10

39/C

1CP2

2651

GView Online / Journal Homepage / Table of Contents for this issue

Page 2: Citethis:Phys. Chem. Chem. Phys.,2012,14 ,675680 PAPER · his ournal is c the Owner Societies 2012 Phys. Chem. Chem. Phys.,2012,14,675680 677 Many more TOF spectra were taken at various

676 Phys. Chem. Chem. Phys., 2012, 14, 675–680 This journal is c the Owner Societies 2012

We find here that both channels 1 and 2 occur with kinetic energy

distributions consistent with statistical decay on the ground state

surface. The branching ratio is about 1.7 (5) in favor of the higher

energy channel 2. These results are interpreted based on simple

RRKM considerations.

II. Experimental

The molecular beam photodissociation apparatus has been

described in previous papers23–25 and is shown in Fig. 2. A fixed

molecular beam is intersected by a pulsed laser and a rotatable

detector is used to analyze the photodissociation products via

electron impact ionization mass spectroscopy. Photofragment

time-of-flight measurements for ion masses of interest are taken

at a variety of laboratory scattering angles Ylab.

In more detail, a pulsed molecular beam of isobutene was

formed using a piezo-activated valve (Physik Intrumente, PI)

at a repetition rate of 200 Hz. The gas mixture was 1%

isobutene (purity greater than 99%, SynQuest Labs Inc, FL)

in helium or neon with a backing pressure of 25 psig. The

molecular beam was collimated with two skimmers, one of

which separated the source chamber from the main chamber

where dissociation and detection occurs. In the main chamber,

the molecular beam was crossed at a 901 by a focused photo-

dissociation beam (2 � 4 mm) produced by an ArF (193 nm)

excimer laser (Lambda-Physik) operating at a repetition rate

of 100 Hz with a typical pulse energy of 40 mJ. The scattered

photofragments were detected in the plane defined by the

molecular and laser beams as a function of the laboratory

angle, Ylab, measured with respect to the molecular beam.

After reaching the triply differentially pumped detection

region, the neutral photofragments were ionized by electron

impact ionization (77 eV), mass-selected using a quadrupole

mass spectrometer, and observed with a Daly style ion detector.

Photofragment time of flight (TOF) spectra with respect to the

laser pulse were obtained by measuring ion signal as a function

of time and were digitally recorded by the use of a multichannel

scaler connected to a PC-computer. Spectra in which helium

was the carrier gas were accumulated over 50 000 laser shots;

the data acquisition time was increased to 2-300 000 laser shots

when Ne was used as the carrier gas. Background subtraction

was performed in order to only observe signal relevant to the

photodissociation of isobutene.

The molecular beam was characterized using a slotted

chopper spinning at a 200 Hz. Characteristic beam velocities

of about 1700 m s�1 were obtained in the present experiment

using He as a seed gas. The corresponding speed ratio (beam

flow velocity over velocity spread) was consistently in the range

of 35–40. When using Ne as a carrier gas, typical beam

velocities were 800 m s�1 with a speed ratio of 20.

III. Results

Fig. 3 shows representative TOF data at m/z = 55 (C4H7+),

the parent ion for the C4H7 fragment that would be formed by

H atom loss via channel 1. Three of these spectra were taken

using He as the seeding gas, while the fourth, at Y = 91, was

taken with Ne as the carrier gas. Fig. 4 shows spectra at

m/z = 41 (C3H5+) and m/z = 15 (CH3

+), corresponding to

the parent ions for channel 2. These spectra were taken with

He as the carrier gas. In all TOF spectra, open circles represent

the data and the solid lines represent the simulated fit to the

data obtained using forward convolution of assumed center-

of-mass translational energy distributions (see Section IV).

Fig. 1 Potential energy surface of isobutene based on ref. 21 and 22.

Fig. 2 Schematic representation of the apparatus, showing the mole-

cular beam source, the photodissociation laser and the rotating mass

spectrometer detector.

Fig. 3 TOF spectra for m/z = 55 (C4H7+) fragments collected at

Ylab = 41, 111 and 131 using He as a seeding gas and 91 using Ne

obtained from 193 nm photodissociation of C4H8. The fits to these

TOF spectra are generated from the P(ET) distribution in Fig. 6.

Dow

nloa

ded

by U

nive

rsity

of

Cal

ifor

nia

- B

erke

ley

on 1

7 Ja

nuar

y 20

12Pu

blis

hed

on 2

8 N

ovem

ber

2011

on

http

://pu

bs.r

sc.o

rg |

doi:1

0.10

39/C

1CP2

2651

G

View Online

Page 3: Citethis:Phys. Chem. Chem. Phys.,2012,14 ,675680 PAPER · his ournal is c the Owner Societies 2012 Phys. Chem. Chem. Phys.,2012,14,675680 677 Many more TOF spectra were taken at various

This journal is c the Owner Societies 2012 Phys. Chem. Chem. Phys., 2012, 14, 675–680 677

Many more TOF spectra were taken at various ion masses and

angles. For example, Fig. 5 shows several TOF spectra taken

at m/z = 39 (C3H3+) using Ne as the carrier gas; these spectra

are particularly useful for determining the product branching

ratio as discussed in Section IV.

The TOF spectra for m/z = 55 (Fig. 3, He and Ne carrier

gas) consist of a single peak whose intensity decreases with

increasing laboratory angle from 41 to 131. No signal was

observed beyond 131, the kinematic limit for H atom loss at

193 nm based on the appropriate Newton diagram. Hence,

since we are only observing a single peak for m/z = 55 that

disappears beyond 131, we can attribute this feature to H atom

loss from the isobutene molecule at 193 nm.While them/z=41

and 15 spectra in Fig. 4 could, in principle, originate from

dissociative ionization of the C4H7 fragment, these spectra

extend over a much larger angular range, indicating they

correspond to at least one additional photodissociation channel,

with channel 2 as the likely candidate. The definitive assignment

of these features is presented in Section IV.

IV. Analysis

The data were analyzed by constructing photofragment center-

of-mass translational energy and angular distributions P(ET, y)for all photodissociation channels and using these to simulate

the laboratory-frame TOF spectra. For each channel, the overall

distribution can be decoupled into a product of center-of-mass

translational energy P(ET) and angular distributions l(y) as follows:

P(ET, y) = P(ET)I(y) (3)

The TOF spectra were fit by forward convolution of center-of-

mass energy and angular distributions using the PHOTRAN

software package.26 In our experimental geometry, an aniso-

tropic angular distribution in the plane of detection can occur

even with an unpolarized laser beam since the rotational axis

of the detector is orthogonal to the plane defined by the

molecular and the laser beam, but a satisfactory fit to all the

data was obtained assuming an isotropic distribution. The

entire set of TOF data could be fit using the P(ET) distributions

in Fig. 6 and 7 for channels 1 and 2, respectively; the simulated

spectra obtained from these distributions are shown as solid

lines in Fig. 3, 4, and 5.

From conservation of energy, the translational energy ET, is

given by:

ET = hn + E0 � Eint � D0 (4)

Here hn is the photon energy at 193 nm,D0 the dissociation energy

of a given photodissociation channel (from eqn (1) and (2)),

Eint is the internal energy of the photofragment and E0 is the

initial energy of the isobutene molecule. Hence, for a given

photodissociation channel of cold molecules (i.e., E0 = 0), the

maximum translational energy ETmax is given by hn � D0,

yielding values of 60 and 47 kcal mol�1 for channels 1 and 2,

respectively, based on the energetics of Fig. 1. This constraint

was applied to the P(ET) distributions used to fit the data;

other than that, the point-wise distributions were adjusted

freely to obtain the best simulation of the full data set.

The P(ET) distribution for channel 1 in Fig. 6 fits the entire

set of TOF spectra at m/z = 55. The distribution peaks at

4 kcal mol�1 with an average translational energy, hETi =11.1 kcal mol�1, and extends up to ETmax = 60 kcal mol�1.

The simulations are not particularly sensitive to products with

Fig. 4 TOF spectra for m/z = 41 (C3H5+) and m/z = 15 (CH3

+)

fragments collected at Ylab = 71, 131 and Ylab = 151, 301 respectively

obtained from 193 nm photodissociation of isobutene(C4H8). A single

P(ET) distribution, shown in Fig. 7, was used to fit these spectra.

Fig. 5 Representative TOF spectra ofm/z=39 (C3H3+) at laboratory

angles ofYlab = 111 and 131. The doted and dashed lines show simulated

TOF spectra using the P(ET) distributions in Fig. 6 and 7 respectively. The

solid line shows the sum of dashed and dotted simulations.

Fig. 6 Center-of-mass P(ET) distribution from isobutene photo-

dissociation at 193 nm to H + C4H7. The maximum available transla-

tional energy is 60 kcal mol�1. Products with ET o 4.5 kcal mol�1 can

undergo secondary dissociation (see text).

Dow

nloa

ded

by U

nive

rsity

of

Cal

ifor

nia

- B

erke

ley

on 1

7 Ja

nuar

y 20

12Pu

blis

hed

on 2

8 N

ovem

ber

2011

on

http

://pu

bs.r

sc.o

rg |

doi:1

0.10

39/C

1CP2

2651

G

View Online

Page 4: Citethis:Phys. Chem. Chem. Phys.,2012,14 ,675680 PAPER · his ournal is c the Owner Societies 2012 Phys. Chem. Chem. Phys.,2012,14,675680 677 Many more TOF spectra were taken at various

678 Phys. Chem. Chem. Phys., 2012, 14, 675–680 This journal is c the Owner Societies 2012

ET o 4 kcal mol�1, a point covered further in Section V, but

the tail extending out to ETmax produces a better fit to our data

than distributions with a lower energy cutoff.

Forward convolution of the P(ET) distribution in Fig. 7

shows that the TOF spectra for m/z = 41 and m/z = 15 in

Fig. 4 correspond to momentum-matched C3H5 and CH3

photofragments from mass channel 2. The P(ET) distribution

for this channel peaks at 4 kcal mol�1 with an average

translational energy of 5.1 kcal mol�1 and extends to up to

17 kcal mol�1. The P(ET) distribution could be extended to

ETmax (47 kcal mol�1) without degrading the quality of the fit.

The analysis shows that contributions to the TOF spectra in

Fig. 4 from dissociative ionization (from m/z = 41 and 55) are

negligible. Fig. 1 shows that that the primary product of the

photodissociation of isobutene via a methyl loss can further

isomerize via a 1,2-hydrogen shift to form the allyl radical,22 a

process considered further in Section V.

The channel 2 : channel 1 branching ratio can be extracted

from TOF spectra at ion masses where both channels con-

tribute via dissociative ionization, such as them/z=39 spectra

shown in Fig. 5. The branching ratio is obtained from eqn (5).

CH3 loss channel

H loss channel¼ R� sC4H7

sC3H5

� fC4H7

fC3H5

ð5Þ

Here, R describes the relative weight of the two P(ET) distri-

butions used to fit the experimental data for both channels in

order to reproduce the relative intensity of each contribution

of the H loss and CH3 loss channel. The relative electron

impact ionization cross section, si, has been determined using

the additivity method proposed by Fitch et al.27 Finally,

f represents the fraction of a given photodissociation channel

signal appearing at m/z= 39 via dissociative ionization. These

fractions, 30% for the H loss channel and 49% for CH3 loss,

were determined by taking TOF spectra at Ylab = 91 at all

values of m/z that yield to a measurable signal (m/z = 55, 54,

53, 52, 51, 50, 49, 41, 40, 39, 38, 37, 36, 27, 26 and 25). Fig. 5

shows two representative TOF spectra for m/z = 39 and

indicates the simulated contributions of dissociative ionization

from channels 1 and 2; the simulations are generated from the

P(ET) distributions in Fig. 6 and 7. By optimizing R to generate

the best agreement with the TOF spectra, we obtained a

branching ratio of 1.7 (5) in favor of CH3 loss. The main source

of uncertainty for the branching ratio comes from the determi-

nation of R which may vary between 1.5 and 2.5 with best

agreement found at 2.1.

Finally, electronic structure calculations were carried out to

characterize the reaction coordinates for channels 1 and 2,

i.e. C–H and C–C bond cleavage. The calculations were

performed at the MP2/6-311+g(d,p) level of theory and basis

set using the GAUSSIAN0328 software package. The reaction

coordinates were mapped out optimizing the geometry as either

rCH or rCC was varied. Calculations were carried out to bond

lengths as large as 3.8 A for both bonds. The optimized bond

lengths were found to be rCH = 1.093 A and rCC = 1.506 A at

the equilibrium geometry. No exit barrier was found for either

channel, as expected for simple bond cleavage. Calculated

energies along the two reaction coordinates were scaled to

match DH0 of eqn (1) and (2); these scaled energies were used

in the RRKM calculations described in Section V. Harmonic

vibrational frequencies were calculated for vibrational modes

perpendicular to both reaction coordinates for use in RRKM

rate constant calculations discussed in Section V.

V. Discussion

The preceding analysis shows that photodissociation of iso-

butene at 193 nm results in both H and CH3 loss as primary

photofragmentation channels. The P(ET) distributions for

both channels in Fig. 6 and 7 peak at low kinetic energy. This

result, in conjunction with our electronic structure calculations

that show no exit barrier for either channel, suggests that the

overall dissociation mechanism involves internal conversion to

the ground state followed by statistical unimolecular decay to

products. If this is the case, however, then we need to be able

to explain why the CH3 loss channel is favored even though it

lies higher in energy by B13 kcal mol�1, assuming that the

other fragment is the 2-propenyl radical. In the remainder of

this section, we consider various aspects of the two channels in

more detail and conclude by discussing calculations of the

branching ratio using RRKM theory.

The P(ET) distribution for channel 1 shown in Fig. 6 peaks

at 4 kcal mol�1. However, as indicated in Fig. 1, the 2-methyl-

allyl radical formed via H atom loss can undergo further

dissociation to CH3 + C3H4; the calculated barrier for this

process is 55.5 kcal mol�1.21 Any C4H7 fragments from

channel 1 with internal energy exceeding this barrier height

can undergo secondary dissociation. Since the H atom has

no internal excitation, the C4H7 internal energy is given by

ETmax � ET, so fragments with ET o 4.5 kcal mol�1 would

have enough internal energy to dissociate. We found that a

P(ET) distribution truncated below 4.5 kcal mol�1 (shown as a

dotted line in Fig. 6) produced fits to our data atm/z= 55 that

are indistinguishable from those generated from the distri-

bution in Fig. 6, so we cannot claim to see clear evidence for

secondary dissociation. However, if this process is occurring, it

only affects a small part of the P(ET) distribution and will not

significantly alter the calculated branching ratio.

Fig. 7 Center-of-mass P(ET) distribution from isobutene photo-

dissociation at 193 nm to CH3 + C3H5. The maximum allowed

translational energy is 47 kcal mol�1.

Dow

nloa

ded

by U

nive

rsity

of

Cal

ifor

nia

- B

erke

ley

on 1

7 Ja

nuar

y 20

12Pu

blis

hed

on 2

8 N

ovem

ber

2011

on

http

://pu

bs.r

sc.o

rg |

doi:1

0.10

39/C

1CP2

2651

G

View Online

Page 5: Citethis:Phys. Chem. Chem. Phys.,2012,14 ,675680 PAPER · his ournal is c the Owner Societies 2012 Phys. Chem. Chem. Phys.,2012,14,675680 677 Many more TOF spectra were taken at various

This journal is c the Owner Societies 2012 Phys. Chem. Chem. Phys., 2012, 14, 675–680 679

For channel 2, the C–C bond fission products are CH3 +

CH3CCH2, the 2-propenyl radical. As seen in Fig. 1, this

species lies 20.6 kcal mol�1 above the allyl radical,22 and one

must then consider the possible role of allyl in the dissociation

dynamics. For example, it is possible for allyl to be formed

directly from the dissociation of isobutene through a multi-

center transition state, in which an H atom from the remaining

CH3 group transfers to the central C atom as the methyl

fragment departs. Such a transition state typically results in a

substantial exit barrier along the reaction coordinate which

would then lead to a translational energy distribution peaking

well away from ET = 0, in contrast to the distribution in

Fig. 7. Hence concerted production of allyl seems unlikely.

It is also possible for 2-propenyl products formed with more

than 45.4 kcal mol�1 internal energy to isomerize to allyl by a

1,2-hydrogen shift, according to the energetics in Fig. 1.

However, since ETmax for channel 2 is only 47 kcal mol�1 at

193 nm, this means that isomerization can only occur for

dissociation events with ET o 1.6 kcal mol�1, which corres-

ponds to a very small fraction of the P(ET) distribution in

Fig. 7. Moreover, this value of ET assumes no internal excita-

tion of the CH3 fragment. Isomerization to allyl therefore

represents a very minor contribution to the overall dynamics

and may not occur at all.

It thus appears that our channel 2 : channel 1 branching ratio of

1.7 (5) favors the higher energy C–C bond fission channel over

C–H bond fission, a somewhat unexpected result at first glance.

However, as discussed in previous work on isobutene kinetics,13,14

the A-factor for CH3 loss is higher than for H loss because CH3

loss results in three more rotational degrees of freedom at the

transition state than H loss. As a result, CH3 loss should become

faster than H loss at a sufficiently high temperature.

We have explored this effect from the perspective of our

experiment by calculating microcanonical rate constants ki(E)

for the two channels with RRKM theory,29

kiðEÞW�ðE � E0Þ

hrðEÞ ð6Þ

Here W* defines the total number of states of the critical

configuration, E0 is the energy of the transition state, and r(E)denotes the density of states of the reactant at total energy E.

The density and sum of states were calculated by direct state-

count method using the Beyer-Swinehart algorithm,30,31 using

vibrational frequencies obtained from the electronic structure

calculations described in Section IV. Vibrational frequencies

for all modes perpendicular to the reaction coordinate were

calculated for both bond fission channels. For rCC 4 2.6 A,

we treated torsional motion of the methyl group as a one-

dimensional free rotor with a rotational constant of around

5 cm�1.

By looking at the evolution of the calculated rate constants

for both channels as a function of fragment separation, we found

minimum values of the rate constants for channels 1 and 2 at

rCH = 3.4 A and rCC = 3.4 A, respectively. These values were

kCH3= 1.55 � 108 s�1 and kH = 8.42 � 107 s�1, leading to a

theoretical branching ratio of CH3 loss :H loss of 1.8, in

remarkably close agreement with experiment. We also calcu-

lated the branching ratio as a function of excitation energy

at these two transition state geometries as shown in Fig. 8. At

193 nm excitation (148 kcal mol�1), methyl loss dominates

over H loss. However, the H loss channel becomes more

important with decreasing photon energy (i.e. increasing wave-

length) to become the dominant channel at about 133 kcal mol�1

(215 nm), which is below the 205 nm onset15 of the isobutene UV

absorption spectrum. In any case, our RRKM results are

certainly consistent with the arguments put forth in kinetics

papers regarding the A-factors for the two bond fission

channels and suggest that both channels should be considered

when constructing kinetic models for isobutene pyrolysis and

oxidation.

VI. Conclusions

We have investigated the photodissociation dynamics of the

isobutene molecule at 193 nm using photofragment transla-

tional spectroscopy. The translational energy distribution and

the product branching ratio between H and CH3 loss were

obtained. The translational energy distributions indicated that

both channels take place via statistical dissociation on the

ground state potential energy surface. The branching ratio

between both channels was determined experimentally to be

1.7(5) in favor of the higher energy CH3 loss channel. Electronic

structure calculations combined with RRKM theory showed

that such a result is consistent with statistical dissociation at the

excitation energy used in our experiment.

Acknowledgements

This work was supported by the Director, Office of Basic Energy

Sciences, Chemical Sciences Division of the U.S. Department of

Energy under Contract No. DE-AC02-05CH11231.

References

1 V. Dias and J. Vandooren, Fuel, 2010, 89, 2633.2 S. Sato and R. J. Cvetanovic, Can. J. Chem., 1958, 36, 970.3 T. Berndt and O. Boge, J. Atmos. Chem., 1995, 21, 275.4 D. E. Woon and J. Y. Park, Icarus, 2009, 202, 642.

Fig. 8 Evolution of the rate constant ratio calculated using RRKM

theory between the CH3 loss channel and the H loss channel at

fragment separation of 3.4 A as a function of energy.

Dow

nloa

ded

by U

nive

rsity

of

Cal

ifor

nia

- B

erke

ley

on 1

7 Ja

nuar

y 20

12Pu

blis

hed

on 2

8 N

ovem

ber

2011

on

http

://pu

bs.r

sc.o

rg |

doi:1

0.10

39/C

1CP2

2651

G

View Online

Page 6: Citethis:Phys. Chem. Chem. Phys.,2012,14 ,675680 PAPER · his ournal is c the Owner Societies 2012 Phys. Chem. Chem. Phys.,2012,14,675680 677 Many more TOF spectra were taken at various

680 Phys. Chem. Chem. Phys., 2012, 14, 675–680 This journal is c the Owner Societies 2012

5 M. J. Nam, S. E. Youn and J. H. Choi, J. Chem. Phys., 2006,124, 8.

6 M. Zierhut, W. Roth and I. Fischer, J. Phys. Chem. A, 2004,108, 8125.

7 B. Negru, G. M. P. Just, D. Park and D. M. Neumark, Phys.Chem. Chem. Phys., 2011, 13, 8180.

8 J. N. Bradley and K. O. West, J. Chem. Soc., Faraday Trans. 1,1976, 72, 558.

9 K. Brezinsky and F. L. Dryer, Combust. Sci. Technol., 1986,45, 225.

10 H. J. Curran, M. P. Dunphy, J. M. Simmie, C. K. Westbrook andW. J. Pitz, Symp. (Int.) Combust., [Proc.], 1992, 24, 769.

11 J. C. Bauge, F. Battin-Leclerc and F. Baronnet, Int. J. Chem.Kinet., 1998, 30, 629.

12 S. Santhanam, J. H. Kiefer, R. S. Tranter and N. K. Srinivasan,Int. J. Chem. Kinet., 2003, 35, 381.

13 K. Yasunaga, Y. Kuraguchi, R. Ikeuchi, H. Masaoka,O. Takahashi, T. Koike and Y. Hidaka, Proc. Combust. Inst.,2009, 32, 453.

14 A. Fahr, J. B. Halpern and D. C. Tardy, J. Phys. Chem. A, 2007,111, 6600.

15 E. P. Carr and H. Stucklen, J. Chem. Phys., 1936, 4, 760.16 M. Caricato, G. W. Trucks, M. J. Frisch and K. B. Wiberg,

J. Chem. Theory Comput., 2010, 6, 370.17 M. H. Palmer, A. J. Beveridge, I. C. Walker and T. M. Abuain,

Chem. Phys., 1987, 117, 51.18 K. B. Wiberg, A. E. de Oliveira and G. Trucks, J. Phys. Chem. A,

2002, 106, 4192.19 P. G. Wenthold, M. L. Polak and W. C. Lineberger, J. Phys.

Chem., 1996, 100, 6920.20 J. L. Miller, J. Phys. Chem. A, 2004, 108, 2268.21 Y. Li, H. L. Liu, Z. J. Zhou, X. R. Huang and C. C. Sun, J. Phys.

Chem. A, 2010, 114, 9496.

22 H. J. Deyerl, I. Fischer and P. Chen, J. Chem. Phys., 1999,110, 1450.

23 Y. T. Lee, J. D. McDonald, P. R. Lebreton and D. R. Herschbach,Rev. Sci. Instrum., 1969, 40, 1402.

24 J. C. Robinson, S. A. Harris, W. Sun, N. E. Sveum and D. M.Neumark, J. Am. Chem. Soc., 2002, 124, 10211.

25 S. J. Goncher, D. T. Moore, N. E. Sveum and D. M. Neumark,J. Chem. Phys., 2008, 128, 114303.

26 S. A. Harich, PHOTRAN, 2003.27 W. L. Fitch and A. D. Sauter, Anal. Chem., 1983, 55, 832.28 M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E. Scuseria,

M. A. Robb, J. R. Cheeseman, G. Scalmani, V. Barone,B. Mennucci, G. A. Petersson, H. Nakatsuji, M. Caricato, X. Li,H. P. Hratchian, A. F. Izmaylov, J. Bloino, G. Zheng,J. L. Sonnenberg, M. Hada, M. Ehara, K. Toyota, R. Fukuda,J. Hasegawa, M. Ishida, T. Nakajima, Y. Honda, O. Kitao,H. Nakai, T. Vreven, J. A. Montgomery, Jr., J. E. Peralta,F. Ogliaro, M. Bearpark, J. J. Heyd, E. Brothers, K. N. Kudin,V. N. Staroverov, R. Kobayashi, J. Normand, K. Raghavachari,A. Rendell, J. C. Burant, S. S. Iyengar, J. Tomasi, M. Cossi,N. Rega, J. M. Millam, M. Klene, J. E. Knox, J. B. Cross,V. Bakken, C. Adamo, J. Jaramillo, R. Gomperts, R. E.Stratmann, O. Yazyev, A. J. Austin, R. Cammi, C. Pomelli,J. W. Ochterski, R. L. Martin, K. Morokuma, V. G. Zakrzewski,G. A. Voth, P. Salvador, J. J. Dannenberg, S. Dapprich, A. D.Daniels, O. Farkas, J. B. Foresman, J. V. Ortiz, J. Cioslowski andD. J. Fox, R. A. Gaussian 09, Gaussian, Inc., Wallingford, CT,2009.

29 R. A. Marcus and O. K. Rice, J. Phys. Chem., 1951, 55, 894.30 T. Beyer and D. F. Swinehart, Commun. ACM, 1973, 16, 379.31 R. G. Gilbert and S. C. Smith, Theory of unimolecular and

recombination reactions, Blackwell Scientific Publications Journal,Oxford, 1990.

Dow

nloa

ded

by U

nive

rsity

of

Cal

ifor

nia

- B

erke

ley

on 1

7 Ja

nuar

y 20

12Pu

blis

hed

on 2

8 N

ovem

ber

2011

on

http

://pu

bs.r

sc.o

rg |

doi:1

0.10

39/C

1CP2

2651

G

View Online